A Bayesian decision analysis approach to assess voice disorder risks by using acoustic features

Vocal fold nodules are recognized as an occupational disease for all collective of workers performing activities for which maintained and continued use of voice is required. Computer‐aided systems based on features extracted from voice recordings have been considered as potential noninvasive and low cost tools to diagnose some voice‐related diseases. A Bayesian decision analysis approach has been proposed to classify university lectures in three levels of risk: low, medium, and high, based on the information provided by acoustic features extracted from healthy controls and people suffering from vocal fold nodules. The proposed risk groups are associated with different treatments. The approach is based on the calculation of posterior probabilities of developing vocal fold nodules and considers utility functions that include the financial cost and the probability of recovery for the corresponding treatment. Maximization of the expected utilities is considered. By using this approach, the risk of having vocal fold nodules is identified for each university lecturer, so he/she can be properly assigned to the right treatment. The approach has been applied to university lecturers according to the Disease Prevention Program of the University of Extremadura. However, it can also be applied to other voice professionals (singers, speakers, coaches, actors…).     

Review of the European Inland Fisheries Advisory Commission (EIFAC) experiments on stock assessment in lakes

A series of international collaborative experiments carried out under the auspices of EIFAC sought to compare various sampling methods for stock abundance estimation in European inland waters. Four water bodies were used: the Evo lakes (Finland), Lake Konnevesi (Finland), Lake Constance (Germany) and LakeTegel (Germany). The experiments emphasized the difficulties in obtaining reliable fish stock estimates when using different sampling methods. Comparisons between the various methods were also limited. Poor data reliability was attributed both to problems inherent in the various methods and to deficiencies in the sampling design. These results emphasize the caution with which sampling data from inland water bodies should be interpreted. Suggestions are made by the EIFAC group for further studies.

Zusammenfassung

Überblick über die EIFAC-Experimente zur Bestandsabschätzung bei Seen
In internationaler Zusammenarbeit wurde eine Serie von Exerimenten auf Anregung der EIFAC durchgeführt. Dabei wurden Methoden der Fischbestandsabsciätzung in europäischen Binnengewässern verglichen. Die Experimente wurden an 4 Orten durchgeführt: den Evo-Seen und dem Konnevesisee in Finnland sowie dem Bodensee und dem Tegeler See in Deutschland. Es zeigte sich die Schwierigkeit, die Ergebnisse verschiedener Methoden zu vergleichen und zu einer befriedienden Bestandsschätzung zu kommen. Die Schwierigkeiten gehen auf die Methoden selbst wie auf Be Art ihres Einsatzes zurück. Die Ergebnisse der Experimente mahnen zur Vorsicht bei der Interpretation von Bestandsabschätzungen in Binnengewässern. Es werden Vorschläge für das kftige Vorgehen der EIFAC-Arbeitsgruppe und anderer Bearbeiter gemacht.     

A Bayesian approach for subgroup analysis

Several penalization approaches have been developed to identify homogeneous subgroups based on a regression model with subject-specific intercepts in subgroup analysis. These methods often apply concave penalty functions to pairwise comparisons of the intercepts, such that the subjects with similar intercept values are assigned to the same group, which is very similar to the procedure of the penalization approaches for variable selection. Since the Bayesian methods are commonly used in variable selection, it is worth considering the corresponding approaches to subgroup analysis in the Bayesian framework. In this paper, a Bayesian hierarchical model with appropriate prior structures is developed for the pairwise differences of intercepts based on a regression model with subject-specific intercepts, which can automatically detect and identify homogeneous subgroups. A Gibbs sampling algorithm is also provided to select the hyperparameter and estimate the intercepts and coefficients of the covariates simultaneously, which is computationally efficient for pairwise comparisons compared to the time-consuming procedures for parameter estimation of the penalization methods (e.g., alternating direction method of multiplier) in the case of large sample sizes. The effectiveness and usefulness of the proposed Bayesian method are evaluated through simulation studies and analysis of a Cleveland Heart Disease Dataset.     

Statistical analysis of microarray data: a Bayesian approach

The potential of microarray data is enormous. It allows us to monitor the expression of thousands of genes simultaneously. A common task with microarray is to determine which genes are differentially expressed between two samples obtained under two different conditions. Recently, several statistical methods have been proposed to perform such a task when there are replicate samples under each condition. Two major problems arise with microarray data. The first one is that the number of replicates is very small (usually 2-10), leading to noisy point estimates. As a consequence, traditional statistics that are based on the means and standard deviations, e.g. t-statistic, are not suitable. The second problem is that the number of genes is usually very large (approximately 10,000), and one is faced with an extreme multiple testing problem. Most multiple testing adjustments are relatively conservative, especially when the number of replicates is small. In this paper we present an empirical Bayes analysis that handles both problems very well. Using different parametrizations, we develop four statistics that can be used to test hypotheses about the means and/or variances of the gene expression levels in both one- and two-sample problems. The methods are illustrated using experimental data with prior knowledge. In addition, we present the result of a simulation comparing our methods to well-known statistics and multiple testing adjustments.     

Fisheries in the Southern Ocean: an ecosystem approach

The Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) is bound by its Article II, 3 to follow an ecosystem approach to management. This approach has been extended to the application of a precautionary approach in the late 1980s. In our review, we deal primarily with the science-related aspects of CCAMLR and its development towards an ecosystem approach to the management of the living resources of the Southern Ocean. To assist the Commission in meeting its objectives, as set out in Article II, 3, the Scientific Committee established the CCAMLR Ecosystem Monitoring Programme to detect possible effects of krill fishing on the performance of top-level predators, such as albatrosses, penguins, petrels and fur seals. Fisheries in the Southern Ocean followed the fate of other fisheries worldwide in which target species were depleted to low level one after the other. Currently, two types of fisheries are open: the longline fisheries on Patagonian toothfish (Dissostichus eleginoides) and Antarctic toothfish (Dissostichus mawsoni) and the trawl fisheries on mackerel icefish (Champsocephalus gunnari). Both fisheries are managed in a single-species context, however, with conservation measures in place to protect by-catch species, such as rattails (Macrouridae) and skates and rays (Rajidae). Two major problems still exist in fisheries in the Southern Ocean: the by-catch of birds in longline fisheries primarily in the Indian Ocean and the high level of IUU fishing again in the Indian Ocean. Both, the by-catch of birds and high IUU catches undermine the credibility of CCAMLR to safeguard the marine living resources in the Southern Ocean.     

A Bayesian decision approach for sample size determination in phase II trials

Stallard (1998, Biometrics 54, 279-294) recently used Bayesian decision theory for sample-size determination in phase II trials. His design maximizes the expected financial gains in the development of a new treatment. However, it results in a very high probability (0.65) of recommending an ineffective treatment for phase III testing. On the other hand, the expected gain using his design is more than 10 times that of a design that tightly controls the false positive error (Thall and Simon, 1994, Biometrics 50, 337-349). Stallard's design maximizes the expected gain per phase II trial, but it does not maximize the rate of gain or total gain for a fixed length of time because the rate of gain depends on the proportion of treatments forwarding to the phase III study. We suggest maximizing the rate of gain, and the resulting optimal one-stage design becomes twice as efficient as Stallard's one-stage design. Furthermore, the new design has a probability of only 0.12 of passing an ineffective treatment to phase III study.     

Microfluorometry analysis II. A Bayesian approach.

In flow microfluorometry (FMF) analysis cells stained with a fluorescent dye that binds specifically to DNA are passed through the instrument. The number of cells in the population having a fluorescence intensity is recorded in a single channel of a multichannel pulse height analyzer. The result is a DNA fluorescence histogram for the population.A method for decomposing an FMF histogram into its G₁, S and G₂ + M components, corresponding to similarly designated phases of the cell cycle is given. This technique can also be applied to find the parameters in all of the previous approaches. The parameters are calculated by iteration which eliminates the need for non-linear optimization procedures.     

A decision-analytic framework for impact assessment part I: LCA and decision analysis

Life-cycle assessments (LCAs) are conducted to satisfy the aspiration of decision makers to consider the environment in their decision making. This paper reviews decision analysis and discusses how it can be used to structure the assessment and to integrate characterization and valuation. The decision analytic concepts of objectives (goals) and attributes (indicators of the degree to which an objective is achieved) are used to describe steps of the assessment of the entire impact chain. Decision analysis distinguishes among different types of objectives and attributes; it describes how these relate to each other. Impact indicators such as the Human Toxicity Potential are constructed attributes. A means-ends objectives network can show how the different constructed attributes relate to the objective of protecting the environment. As LCA takes disparate environmental impacts into account, it needs to assess their relative importance. Trade-off methods in decision analysis are grouped into utility theory and multicriteria decision aids; they have different advantages and disadvantages, but are all more sophisticated than simple weighting. The performance of the different trade-off methods has not yet been tested in an LCA context. In the second part of the paper, we present criteria for the development of characterization methods.     

BGX: a fully Bayesian integrated approach to the analysis of Affymetrix GeneChip data

We present Bayesian hierarchical models for the analysis of Affymetrix GeneChip data. The approach we take differs from other available approaches in two fundamental aspects. Firstly, we aim to integrate all processing steps of the raw data in a common statistically coherent framework, allowing all components and thus associated errors to be considered simultaneously. Secondly, inference is based on the full posterior distribution of gene expression indices and derived quantities, such as fold changes or ranks, rather than on single point estimates. Measures of uncertainty on these quantities are thus available. The models presented represent the first building block for integrated Bayesian Analysis of Affymetrix GeneChip data: the models take into account additive as well as multiplicative error, gene expression levels are estimated using perfect match and a fraction of mismatch probes and are modeled on the log scale. Background correction is incorporated by modeling true signal and cross-hybridization explicitly, and a need for further normalization is considerably reduced by allowing for array-specific distributions of nonspecific hybridization. When replicate arrays are available for a condition, posterior distributions of condition-specific gene expression indices are estimated directly, by a simultaneous consideration of replicate probe sets, avoiding averaging over estimates obtained from individual replicate arrays. The performance of the Bayesian model is compared to that of standard available point estimate methods on subsets of the well known GeneLogic and Affymetrix spike-in data. The Bayesian model is found to perform well and the integrated procedure presented appears to hold considerable promise for further development.     

Fisheries assessment of Chwaka Bay (Zanzibar) – following a holistic approach

This study investigates the typical East African multispecies fishery of Chwaka Bay (Zanzibar) by assessing growth, mortality, exploitation and stock size of six key target species of the fishery. Analyses are based on monthly length–frequency data collected from January to June and September to December in 2014 from the three main landing sites surrounding the bay. Estimated von Bertalanffy growth parameters using the ELEFAN I routine as implemented in FiSAT II varied among species from 27.6 to 64.4 cm (L_∞) and 0.24 to 0.8 per year (K). Current exploitation rates compared to biological reference points calculated from yield‐per‐recruit analysis for five of the six key species indicate that Siganus sutor, Lethrinus borbonicus and Lethrinus lentjan are harvested beyond maximum sustainable levels (E_max). While juvenile retention rates of these three species are >80% and current L_c‐values are lower than L_opt, fishing mortalities for Siganus sutor and Lethrinus borbonicus are highest for specimens above length at first maturity. Two management measures are discussed: (i) an increase in mesh size; and (ii) closure of the destructive dragnet fishery. The first option seems feasible only if the radius of the fishery were increased to capture larger specimens outside the shallow bay area. The second option would leave approximately 550 fishermen unemployed. Reallocating dragnet fishing effort to other gears would lead to a substantial increase in the number of boats, which might create use conflicts over the limited fishing area. The general status of the fishery in Chwaka Bay is considered as representing a ‘full exploitation to over‐exploitation scenario’, with no scope for expansion.     

A Bayesian approach to copy-number-polymorphism analysis in nuclear pedigrees

Segmental copy-number polymorphisms (CNPs) represent a significant component of human genetic variation and are likely to contribute to disease susceptibility. These potentially multiallelic and highly polymorphic systems present new challenges to family-based genetic-analysis tools that commonly assume codominant markers and allow for no genotyping error. The copy-number quantitation (CNP phenotype) represents the total number of segmental copies present in an individual and provides a means to infer, rather than to observe, the underlying allele segregation. We present an integrated approach to meet these challenges, in the form of a graphical model in which we infer the underlying CNP phenotype from the (single or replicate) quantitative measure within the analysis while assuming an allele-based system segregating through the pedigree. This approach can be readily applied to the study of any form of genetic measure, and the construction permits extension to a wide variety of hypothesis tests. We have implemented the basic model for use with nuclear families, and we illustrate its application through an analysis of the CNP located in gene CCL3L1 in 201 families with asthma.     

Hydrological attributes and rheophilic freshwater fish: stock assessment

The mathematical and statistical advances in fitting stock assessment models enabled the emergence of the paradigm of “integrated analysis”, which fits all available data to a single model of population dynamics that traditionally has total catch as the only forcing function of the system. This approach, however, allowed us to include, in a flexible way, the effect of hydrological regime as an additional forcing function. We tried to achieve this flexibility by making the annual recruitment rates and spawning biomass adjustable to the attributes of the hydrological cycle data. Our models showed that these attributes are influential in the population dynamics of Brycon hilarii of the Northern Pantanal, and their inclusion in the models allowed best partial fits (which considered fits only to the data components length- and age-compositions, CPUE of juveniles and adults) than the Base-case (without hydrological attributes). The best partial fits where obtained when the attributes “delay of floods” and “intensity of floods” were forcing the spawning biomass and the annual recruitment respectively, indicating that these characteristics of the population may be influenced by specific attributes of the water level. The use of integrated modeling contributed with the advancement of population ecology knowledge of rheophilic fish. It is recommended that freshwater fisheries management be integrated into the hydrology management.     

Population dynamics and potential of fisheries stock enhancement: practical theory for assessment and policy analysis

The population dynamics of fisheries stock enhancement, and its potential for generating benefits over and above those obtainable from optimal exploitation of wild stocks alone are poorly understood and highly controversial. I review pertinent knowledge of fish population biology, and extend the dynamic pool theory of fishing to stock enhancement by unpacking recruitment, incorporating regulation in the recruited stock, and accounting for biological differences between wild and hatchery fish. I then analyse the dynamics of stock enhancement and its potential role in fisheries management, using the candidate stock of North Sea sole as an example and considering economic as well as biological criteria. Enhancement through release of recruits or advanced juveniles is predicted to increase total yield and stock abundance, but reduce abundance of the naturally recruited stock component through compensatory responses or overfishing. Economic feasibility of enhancement is subject to strong constraints, including trade-offs between the costs of fishing and hatchery releases. Costs of hatchery fish strongly influence optimal policy, which may range from no enhancement at high cost to high levels of stocking and fishing effort at low cost. Release of genetically maladapted fish reduces the effectiveness of enhancement, and is most detrimental overall if fitness of hatchery fish is only moderately compromised. As a temporary measure for the rebuilding of depleted stocks, enhancement cannot substitute for effort limitation, and is advantageous as an auxiliary measure only if the population has been reduced to a very low proportion of its unexploited biomass. Quantitative analysis of population dynamics is central to the responsible use of stock enhancement in fisheries management, and the necessary tools are available.     

A Thurstonian model for quantitative genetic analysis of ranks: a Bayesian approach

A fully Bayesian method for quantitative genetic analysis of data consisting of ranks of, e.g., genotypes, scored at a series of events or experiments is presented. The model postulates a latent structure, with an underlying variable realized for each genotype or individual involved in the event. The rank observed is assumed to reflect the order of the values of the unobserved variables, i.e., the classical Thurstonian model of psychometrics. Parameters driving the Bayesian hierarchical model include effects of covariates, additive genetic effects, permanent environmental deviations, and components of variance. A Markov chain Monte Carlo implementation based on the Gibbs sampler is described, and procedures for inferring the probability of yet to be observed future rankings are outlined. Part of the model is rendered nonparametric by introducing a Dirichlet process prior for the distribution of permanent environmental effects. This can lead to potential identification of clusters of such effects, which, in some competitions such as horse races, may reflect forms of undeclared preferential treatment.     

A Bayesian approach for the analysis of panel-count data with dependent termination

We consider modeling and Bayesian analysis for panel-count data when the termination time for each subject may depend on its history of the recurrent events. We propose a fully specified semiparametric model for the joint distribution of the recurrent events and the termination time. For this model, we provide a natural motivation, derive several novel properties, and develop a Bayesian analysis based on a Markov chain Monte Carlo algorithm. Comparisons are made to other existing models and methods for panel-count data. We demonstrate the usefulness of our new models and methodologies through the reanalysis of a data set from a clinical trial.     

Refining gene signatures: a Bayesian approach

Background  

In high density arrays, the identification of relevant genes for disease classification is complicated by not only the curse of dimensionality but also the highly correlated nature of the array data. In this paper, we are interested in the question of how many and which genes should be selected for a disease class prediction. Our work consists of a Bayesian supervised statistical learning approach to refine gene signatures with a regularization which penalizes for the correlation between the variables selected.